Autonomous welding robots

Abstract

In various examples, a computer-implemented method of generating instructions for a welding robot. The computer-implemented method comprises identifying an expected position of a candidate seam on a part to be welded based on a Computer Aided Design (CAD) model of the part, scanning a workspace containing the part to produce a representation of the part, identifying the candidate seam on the part based on the representation of the part and the expected position of the candidate seam, determining an actual position of the candidate seam, and generating welding instructions for the welding robot based at least in part on the actual position of the candidate seam.

What is claimed is: 
     
       1. A computer-implemented method of generating instructions for a welding robot, the computer-implemented method comprising
 identifying an expected position of a candidate seam on a part to be welded based on a Computer Aided Design (CAD) model of the part, the CAD model includes an annotation of the candidate seam; 
 scanning a workspace containing the part to produce a representation of the part; 
 identifying the candidate seam on the part based on the representation of the part and the expected position of the candidate seam; determining an actual shape of the candidate seam; and 
 generating welding instructions for a welding robot based at least in part on the actual shape of the candidate seam. 
 
     
     
       2. The computer-implemented method of  claim 1 , further comprising:
 determining an expected shape of the candidate seam using the CAD model of the part; 
 determining an actual position of the candidate seam; and 
 determining the actual shape based at least in part on the actual position and the expected shape of the candidate seam. 
 
     
     
       3. The computer-implemented method of  claim 2 , wherein the welding instructions are generated based at least in part on the actual shape. 
     
     
       4. The computer-implemented method of  claim 2 , wherein determining the actual position of the candidate seam includes updating the expected position of the candidate seam based on the representation of the part. 
     
     
       5. The computer-implemented method of  claim 1 , wherein the expected position of the candidate seam is identified based on the annotation. 
     
     
       6. The computer-implemented method of  claim 1 , wherein identifying the candidate seam includes:
 verifying whether the candidate seam is a seam; and 
 identifying, using a neural network, the candidate seam as a type of seam. 
 
     
     
       7. The computer-implemented method of  claim 1 , further comprising executing the welding instructions to cause the welding robot to perform one or more welding operations. 
     
     
       8. A computer-implemented method of generating welding instructions for a welding robot, the computer-implemented method comprising:
 determining a first position of a candidate seam on a first plurality of points representing a weldable object; 
 generating a second plurality of points representing the weldable object, wherein the first plurality of points and the second plurality of points originate from different sources; 
 identifying a second position of the candidate seam on the weldable object using the second plurality of points and the first position; 
 deforming a subset of the first plurality of points in accordance with a subset of the second plurality of points; and 
 generating welding instructions for a welding robot based at least in part on the second position of the candidate seam. 
 
     
     
       9. The computer-implemented method of  claim 8 , further comprising obtaining, via a sensor, image data of the weldable object. 
     
     
       10. The computer-implemented method of  claim 9 , wherein the first plurality of points originates from a Computer Aided Design (CAD) model of the weldable object and the second plurality of points is generated using the image data. 
     
     
       11. The computer-implemented method of  claim 10 , wherein the CAD model includes an annotation of the candidate seam. 
     
     
       12. The computer-implemented method of  claim 11 , wherein the first position of the candidate seam is identified based on the annotation. 
     
     
       13. A robotic welding system comprising:
 a robot controller configured to:
 identify an expected position of a candidate seam on a part to be welded based on a Computer Aided Design (CAD) model of the part, the CAD model includes an annotation of the candidate seam; 
 scan a workspace containing the part to produce a representation of the part; 
 identify the candidate seam on the part based on the representation of the part and the expected position of the candidate seam; 
 determine an actual shape of the candidate seam; and 
 generate welding instructions for a welding robot based at least in part on the actual shape of the candidate seam. 
 
 
     
     
       14. The robotic welding system of  claim 13 , wherein the robot controller is further configured to:
 determine an expected shape of the candidate seam using the CAD model of the part; 
 determine an actual position of the candidate seam; and 
 determine the actual shape based at least in part on the actual position and the expected shape of the candidate seam. 
 
     
     
       15. The robotic welding system of  claim 14 , wherein the robot controller the robot controller is further configured to generate the welding instructions based at least in part on the actual shape. 
     
     
       16. The robotic welding system of  claim 14 , wherein the robot controller is configured to determine the actual position of the candidate seam by updating the expected position of the candidate seam based on the representation of the part. 
     
     
       17. The robotic welding system of  claim 13 , wherein the robot controller is configured to identify the expected position of the candidate seam based on the annotation. 
     
     
       18. The robotic welding system of  claim 13 , wherein, to identify the candidate seam, the robot controller is configured to:
 verify whether the candidate seam is a seam; and 
 identify, using a neural network, the candidate seam as a type of seam. 
 
     
     
       19. The robotic welding system of  claim 13 , further comprising:
 the welding robot, and 
 wherein the robot controller is further configured to execute the welding instructions to cause the welding robot to perform one or more welding operations.